1. Field of the Invention
This invention relates generally to computing systems, and, more particularly, to processing tasks with failure recovery in a computer system.
2. Description of the Related Art
FIG. 1A illustrates an exemplary computer system 100. The computer system 100 includes a processor 102, a north bridge 104, memory 106, Advanced Graphics Port (AGP) device 108, a network interface card (NIC) 109, a Peripheral Component Interconnect (PCI) bus 110, a PCI connector 111, a south bridge 112, a battery 113, an AT Attachment (ATA) interface 114 (more commonly known as an Integrated Drive Electronics (IDE) interface), an SMBus 115, a universal serial bus (USB) interface 116, a Low Pin Count (LPC) bus 118, an input/output controller chip (SuperI/O™) 120, and BIOS memory 122. It is noted that the north bridge 104 and the south bridge 112 may include only a single chip or a plurality of chips, leading to the collective term “chipset.” It is also noted that other buses, devices, and/or subsystems may be included in the computer system 100 as desired, e.g. caches, modems, parallel or serial interfaces, SCSI interfaces, etc.
The processor 102 is coupled to the north bridge 104. The north bridge 104 provides an interface between the processor 102, the memory 106, the AGP device 108, and the PCI bus 110. The south bridge 112 provides an interface between the PCI bus 110 and the peripherals, devices, and subsystems coupled to the IDE interface 114, the SMBus 115, the USB interface 116, and the LPC bus 118. The battery 113 is shown coupled to the south bridge 112. The Super I/O™ chip 120 is coupled to the LPC bus 118.
The north bridge 104 provides communications access between and/or among the processor 102, memory 106, the AGP device 108, devices coupled to the PCI bus 110, and devices and subsystems coupled to the south bridge 112. Typically, removable peripheral devices are inserted into PCI “slots,” shown here as the PCI connector 111, that connect to the PCI bus 110 to couple to the computer system 100. Alternatively, devices located on a motherboard may be directly connected to the PCI bus 110. The SMBus 115 may be “integrated” with the PCI bus 110 by using pins in the PCI connector 111 for a portion of the SMBus 115 connections.
The south bridge 112 provides an interface between the PCI bus 110 and various devices and subsystems, such as a modem, a printer, keyboard, mouse, etc., which are generally coupled to the computer system 100 through the LPC bus 118, or one of its predecessors, such as an X-bus or an Industry Standard Architecture (ISA) bus. The south bridge 112 includes logic used to interface the devices to the rest of computer system 100 through the IDE interface 114, the USB interface 116, and the LPC bus 118. The south bridge 112 also includes the logic to interface with devices through the SMBus 115, an extension of the two-wire inter-IC bus protocol.
FIG. 1B illustrates certain aspects of the south bridge 112, including reserve power by the battery 113, so-called “being inside the RTC (real time clock) battery well” 125. The south bridge 112 includes south bridge (SB) RAM 126 and a clock circuit 128, both inside the RTC battery well 125. The SB RAM 126 includes CMOS RAM 126A and RTC RAM 126B. The RTC RAM 126B includes clock data 129 and checksum data 127. The south bridge 112 also includes, outside the RTC battery well 125, a CPU interface 132, power and system management units 133, and various bus interface logic circuits 134.
Time and date data from the clock circuit 128 are stored as the clock data 129 in the RTC RAM 126B. The checksum data 127 in the RTC RAM 126B may be calculated based on the CMOS RAM 126A data and stored by BIOS during the boot process, such as is described below, e.g. block 148, with respect to FIG. 2. The CPU interface 132 may include interrupt signal controllers and processor signal controllers.
FIG. 2 illustrates a prior art remote management configuration for the computer system 100. A motherboard 101 provides structural and base electrical support for the south bridge 112, the PCI bus 110, the PCI connector 111, the SMBus 115, and sensors 103A and 103B. The NIC 109, a removable add-in card, couples to the motherboard 101, the PCI bus 110, and the SMBus 115 through the PCI connector 111. The NIC 109 includes an Ethernet controller 105 and an ASF microcontroller 107. The Ethernet controller 105 communicates with a remote management server 90, passing management data and commands between the ASF microcontroller 107 and the remote management server 90. The remote management server 90 is external to the computer system 100.
An industry standard specification, generally referred to as the Alert Standard Format (ASF) Specification, defines one approach to “system manageability” using the remote management server 90. The ASF Specification defines remote control and alerting interfaces capable of operating when an operating system of a client system, such as the computer system 100, is not functioning. Generally, the remote management server 90 is configured to monitor and control one or more client systems. Typical operations of the ASF alerting interfaces include transmitting alert messages from a client to the remote management server 90, sending remote control commands from the remote management server 90 to the client(s) and responses from the client(s) to the remote management server 90, determining and transmitting to the remote management server 90 the client-specific configurations and assets, and configuring and controlling the client(s) by interacting with the operating system(s) of the client(s). In addition, the remote management server 90 communicates with the ASF NIC 109 and the client(s)' ASF NIC 109 communicates with local client sensors 103 and the local client host processor.
When the client has an ACPI-aware operating system functioning, configuration software for the ASF NIC 109 runs during a “one good boot” to store certain ASF, ACPI, and client configuration data.
The transmission protocol in ASF for sending alerts from the client to the remote management server 90 is the Platform Event Trap (PET). A PET frame consists of a plurality of fields, including GUID (globally unique identifier), sequence number, time, source of PET frame at the client, event type code, event level, sensor device that caused the alert, event data, and ID fields.
Many events may cause an alert to be sent. The events may include temperature value over or under a set-point, voltage value over or under a set-point, fan actual or predicted failure, fan speed over or under a set-point, and physical computer system intrusion. System operation errors may also be alerts, such as memory errors, data device errors, data controller errors, CPU electrical characteristic mismatches, etc. Alerts may also correspond to BIOS or firmware progression during booting or initialization of any part of the client. Operating system (OS) events may also generate alerts, such as OS boot failure or OS timeouts. The ASF Specification provides for a “heartbeat” alert with a programmable period typically one minute but not to exceed 10 minutes, when the client does not send out the heartbeat, or “I am still here,” message.
Client control functions are implemented through a remote management and control protocol (RMCP) that is a user datagram protocol (UDP) based protocol. RMCP is used when the client is not running the operating system. RMCP packets are exchanged during reset, power-up, and power-down cycles, each having a different message type. The remote management server 90 determines the ASF-RMCP capabilities of the client(s) by a handshake protocol using a presence-ping-request that is acknowledged by the client(s) and followed-up with a presence-pong that indicates the ASF version being used. The remote management server 90 then sends a request to the client to indicate the configuration of the client, which the client acknowledges and follows with a message giving the configuration of the client as stored in non-volatile memory during the “one good boot.” The RMCP packets include a contents field, a type field, an offset field, and a value field.
RMCP message transactions involve a request from the remote management server 90, a timed wait for an acknowledgement followed by a second timed wait for a response. If either of the time limits for the acknowledgement or the response is exceeded, then the remote management server 90 knows that either the client needs some of the packets resent or the client has lost contact due to failure of either the client or the communications link.
The ASF NIC 109 must be able to report its IP (Internet protocol) address (or equivalent) without the intervention of the operating system. Thus, the ASF NIC 109 should be able to receive and reply to ARP (Address Resolution Protocol) requests with the operating system, not interfere with ARP packets when the operating system is running, and wake-up for ARP packets when configured to do so. Note that ACPI includes waking-up for ARP packets as a standard configuration.
The following information is sent to the remote management server 90 from the client as an indication of the configuration of the client: an ACPI description table identifying sensors and their characteristics, ASF capabilities and system type for PET messages, and the client's support for RMCP and the last RCMP command; how the client configures an optional operating system boot hang failure recovery timer; and the SMBIOS identification of the UUID/GUID for PET messages. ASF objects follow the ASL (ASF Software Language) naming convention of ACPI.
Based in part on the above-described features, modem computer systems are becoming more and more robust than their predecessors. Computer systems today process a fairly large number of tasks at any given time. As the number of tasks that are processed increases, the likelihood that some of these tasks may not successful complete (because of errors, for example) also increases. Errant or hung tasks, for example, may adversely affect the performance of the computer system. As such, recovery from these failed tasks is desirable.